- Email: [email protected]
Use of chromatofocusing for separation of β-lactamases
Journal of Chromatography, Elsevier Science Publishers CHROM.
403 (1987) 217-224 B.V., Amsterdam -
Printed
in The Netherlands
19 586
USE OF CHROMATOFOCUSING VIII*. ANALYTICAL PHALOSPORINASES
OF b-LACTAMASES
CHROMATOFOCUSING OF CHROMOSOMAL FROM FOUR KLEBSIELLA STRAINS
SUSANNE GAL, ANDREA TAR, NEZ* and FERENC J. HERNADI Chemotherapy Section, H-4012 (Hungary)
FOR SEPARATION
Department
MAGDOLNA of Pharmacology,
FROMMER-FILEP, University
Medical
BELA
CE-
L. TOTH-MARTI-
School of Dehrecen,
Debrecen
and LkZLt)
KISS
Department (Received
of Biochemistry, March
28th,
L. Kossuth
University,
Debrecen
H-4010
(Hungary)
1987)
SUMMARY
Although still there are Klebsiella strains which do not harbour plasmids and produce constitutive chromosomal p-lactamases, recently clinical isolates were found in ever increasing numbers carrying mainly TEM-, CARB- and OXA type R-factors. We selected four chromosomal cephalosporinase producing Klebsiella strains to study the p1 values of the enzymes and their simultaneous separability from accompanying proteins by chromatofocusing techniques. We compared pl values of the pure and the crude preparations: K. pneumoniae Kl SC 10436: pl,,,, = 6.4, 6.42; K. aerogenes Kl 1082 E: ~1~,,,, = 6.5, pz_& = 6.5; K. oxytoca 1082 p&rude = = 6.42, PIcrude = 6.4; K. oxytoca 20: pl,,,, = 7.62, pIcrude = 7.6. Excellent E: P& agreement of the p1 values among each other, but occasional differences with those obtained by analytical isoelectrofocusing are attributed to methodological diversities and to the presence of satellite enzymes, known to exist in Klebsiella.
INTRODUCTION
Klebsiella is one of the most commonly distributed bacterium species all over the world. Many Klebsieila strains do not harbour plasmids and produce constitutive chromosomal /%lactamases. Most of these /?-lactamases preferentially hydrolyze cephalosporins, including many of those that are resistant to plasmid-coded enzymes. Sawai et al.’ distinguished two major groups of Klebsiella enzymes of broad substrate specificity: (1) chromosomally coded cephalosporinases that hydrolyze benzylpenil For Part VII, see A. Tar, S. Gil, B. L. Toth-Martinez, 368 (1986) 427.
F. J. Hernhdi
0021-9673/87/$03.50
B.V.
0
1987 Elsevier
Science Publishers
and L. Kiss, J. Chromatogr.,
218
S. GAL et al.
cillin, ampicillin and carbenicillin as well, e.g., K. oxytoca FOU 1, FOU 2, FOU 3, FOU 13, pZ = 5.52; R 700, pZ = 5.72; R 30, pZ = 5.82 or 6.03,4; FOU 11, pZ = 6.02; FOU 12, R 210, D 488, pZ = 6.32; 3859, pZ = 6.65; 22264, pZ between 7.9 and 8.26; K. aerogenes K 1, pZ = 6.657; K. pneumoniae 1103, pZ = 7.12~s; (2) chromosomally determined typical cephalosporinases that have little or no activity against the above penicillins and have pZ values above 7.06,9+12. Generally speaking none of these enzymes is identical with the plasmid-mediated fi-lactamases, with one exception’j that is indistinguishable by kinetic analysis and isoelectric focusing from the plasmid SHV-1 p-lactamase (TEM-1 Type 2 or PIT 2, pZ = 7.7; sulfhydryl variable, Temoniera is the name of a patient and PIT is from Professor J. S. Pitton). This is the variant of SHV-1 incorporated into a non-transmissible replicon, probably the bacterial chromosome. Nugent and Hedges13 proposed that the SHV-1 B-lactamase gene evolved as a chromosomal gene in Klebsiella spp. and was later incorporated into a plasmid. Recently Arakawa et al. I4 found that the chromosomally encoded fi-lactamase gene of K. pneumoniae LEN-l (03:Kl) and the TEM_enzyme of Tn3 (transposon) show 67% homology of the deduced amino acid sequence, suggesting that the latter originates from K. pneumoniae. Recently, clinical K. pneumoniae and mainly K. oxytoca isolates have been found to show different modality of resistance related to their broad-spectrum /$ lactamases. These enzymes are produced in high yield by these Klebsiella strains together with more or less chromosomal enzymes 2,5,6. No doubt, these strains harbour different plasmids as well, solely or in combination as reported in some cases: K. oxytoca 22812, 22820, 23659 harbour TEM-1, pZ = 5.4 with high chromosomal enzyme of pZ7.9-8.26; K. pneumoniae S6 and S7 have SHV-1 and SHV-1 + TEM-1 /?-lactamases, respectively15, 22505 has enzymes of pZ 6.5, 7.2, 7.6 and 8.1 and TEM-1 6. HMS-1 of pZ = 5.0 was also found in Klebsieh strains16. As far as this third group is concerned, the following multiple extrachromosomal #?-lactamases have been reported in various Klebsielia strains’ 7: TEM-1, TEM-2, pI = 5.6; CARB-1, pZ 2 5.3; CARB-2, pZ = 5.7; CARB-3, pZ = 5.75; CARB-4, pZ = 4.3; OXA-1, pZ = 7.4; OXA-2, pZ of 7.4 alid 7.7; OXA-3, pZ = 7.0; OXA-4, pZ between 7.45 and 7.5; OXA-5, pZ = 7.6; OXA-6, pZ = 7.8; OXA-7, pZ = 7.65. Another characteristic of these enzymes is an extended spectrum to new plactams of the third generation and azthreonam by mutationle. This was encountered in K. orizae: adaptation of the plasmid SHV-1 resulted in the form&on of SHV-2, pZ = 7.7 (unchanged). Plasmid P45319 carries the transposon coding for SHV-113; the mutant plasmid, however, is called pBP6017. The isoelectric points of enzymes from four Klebsiella strains, producing constitutive chromosomal fi-lactamases in high yield, were examined in our analytical chromatofocusing system and compared with those obtained by analytical isoelectrofocusing (AIEF). The enzymes were also characterized by their separability from accompanying proteins. EXPERIMENTAL
Bacterial strains
Constitutively producing strains of K. oxytoca 1082 E and 20 were kindly provided by Dr. L. R. Then, Department of Pharmaceutical Research, F. Hoff-
CHROMATOFOCUSING
OF b-LACTAMASES.
VIII.
219
mann-La Roche, Basle, Switzerland, K. oxyloca 20 by Dr. C. E. Cooper, Chemotherapeutic Research Centre, Brockham Park, Betchworth, U.K., K. aerogenes Kl 1082 E by Dr. I. N. Simpson, Glaxo Group Research, Greenford, U.K. and by Dr. S. G. Waley, Sir William Dunn School of Pathology, University of Oxford, Oxford, U.K. K. pneumoniae Kl SC 10436 by Dr. K. Bush, The Squibb Institute for Medical Research, Princeton, NJ, U.S.A. Details of the culturing procedure have been published elsewhereZo. PuriJication of B-lactamases Except for K. oxytoca 20, the p-lactamases were purified by affinity chromatography on phenylboronic acid-agarose (hydrophobic spacer arm, Type B) according to Cartwright and Waley 21. The K. oxytoca 20 fi-lactamase showed an unusual elution profile (Fig. Id) on the affinity column: a major fraction was eluted first with 20 mM triethanolamine hydrochloride-0.5 A4 sodium chloride, pH 7.0 washing buffer and another fraction was obtained with 0.5 M borate-0.5 M sodium chloride, pH 7.0 elution buffer. This was an indication that the Type B affinity material was not suitable for quantitative purification of this enzyme. A similar retarded elution was encountered by Cartwright and Waley2’ in the case of K. aerogenes Kl 1082 E on an hydrophylic Type L boronic acid column. We used a classical prepurification method for the K. oxytoca 20 enzyme: the crude enzyme was applied to Sephadex G-75 (20 cm x 3.5 cm) and eluted by 0.05 M phosphate, pH 7.0; after pooling, the active fractions were chromatographed on QAE-Sephadex A-50 (20 cm x 3.5 cm) using 0.025 M Tris-HCl buffer, pH 7.5. A portion of this purified enzyme was used for chromatofocusing as described below. Crude enzyme fractions were made by freezing-thawing and by sonication of resuspended washed cells in 0.05 M phosphate buffer, pH 7.0. Following treatments with nuclease, supernatants were prepared by ultracentrifugation at 105 000 g for 1 h and extensively dialyzed against the above buffer. Chromatofocusing, enzyme assay, protein determination and sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) A Pharmacia C lo/20 column (20 cm x 1 cm) packed with PBE 94 chromatofocusing material was equilibrated with 0.025 A4 imidazole hydrochloride buffer, pH 7.4 for K. pneumoniae Kl SC 10436, K. aerogenes Kl 1082 E and K. oxytoca 1082 E strains or with 0.025 M ethanolamine-acetic acid buffer, pH 9.4 for K. oxytoca 20. For the first three cephalosporinases, the eluent was Polybuffer 74 diluted eight-fold in distilled water and adjusted to pH 4.0 with hydrochloric acid, in the case of K. oxytoca 20, the eluent was Polybuffer 96 diluted ten-fold in distilled water and adjusted to pH 6.0 with acetic acid. fi-Lactamase activity was assayed by measuring the absorbance of Nitrocefin at 486 nm in a cell of 1 cm pathlength as described by O’Callaghan et aZ.22.One unit is that amount of enzyme which is able to hydrolyze 1 ymol of Nitrocefin in 1 min at 37°C. The protein content of the fractions was estimated by measuring their absorbance at 280 nm or as described by Lowry et a1.23. SDS-PAGE was performed according to Weber and 0sborn24.
S. GAL (a)
U.fr.-’
1
1
U.fr._’
CC)
OD2ao
er al.
0.30
(b)
ODzso
Cd)
\; / 0.10 -
M.
J, I c
/
- 0.25
/I I
5
i0
15
fr. no.
Fig. 1. Elution profiles of the crude extracts of /?-lactamases of four Klebsiella strains on Type B phenylboronic acid-agarose affinity columns. Crude extract samples, dialyzed against 20 mM triethanolamine hydrochloride-O.5 M sodium chloride, pH 7.0, were applied; the columns were washed with the same buffer until protein-free fractions appeared. The enzymes were eluted with 0.5 M berate-0.5 M sodium chloride, pH 7.0, elution buffer. Protein (. . .) and activity (- - -) monitoring were as described in the Experimental. The amount of each enzyme and the volumes of the loading buffer are given in Results and Discussion. Fractions eluted were 1.5 ml. (a) K. pneumoniue K1 SC 10436; (b) K. aerogenes Kl 1082 E; (c) K. oxytoca 1082 E; (d) K. oxytoca 20. fr.no. = Fraction number; U.fr.-’ = units per fraction.
RESULTS
AND
DISCUSSION
Fig. 1 shows the results of the purification of Klehsiella p-lactamases on Type B phenylboronic acid-agarose affinity material. K. pneumoniae Kl SC 10436 (a), K. aerogenes Kl 1082 E (b) and K. oxytoca 1082 E (c) had low protein contents and high activities, while K. oxytoca 20 had about the same protein content but low activity (d). When 128 units of K. pneumoniae Kl SC 10436 crude enzyme were loaded onto a column (3 cm x 0.9 cm) in 1.5 ml loading buffer, no detectable amount of activity passed through the column either on loading or upon washing (Fig. la). The enzyme was eluted in two fractions of 3 ml, yielding 121 units or 95% recovery. Under similar conditions, 165 units of crude K. aerogenes Kl 1082 E enzyme were loaded and 161 units, i.e., 98% recovery, were eluted in two fractions of 3 ml (Fig. 1.b). Of 115.5 units of crude K. oxytoca 1082 E enzyme loaded 112.5 units, 96% recovery, were eluted in 3 ml (Fig. lc). The high purification capacity of the Type B affinity material with the above enzymes is noteworthy. When 5.4 units of crude K. oxytoca 20 (Fig. Id) enzyme in 6 ml were loaded on the affinity column, only 4.3 units, 80%, were eluted. The washing buffer eluted about three quarters of the enzyme in the first ten eluate fractions, i.e., clearly retar-
CHROMATOFOCUSING
10
20
OF p-LACTAMASES.
30 (b)
lb
40 U.fr._’
A II 1I I I
5.
4 0480
i0
I
i0 Cd)
pH
4.0 U*fr.-’
- 0.3
I
I
I I
I I
I:\
VIII.
I
’
;
I
I
’
- 0.2
. 0.1 I I
IO
20
30
40
fr.no.
.^
1”
I 1
zu
. .. .. . . . . . .. .
30
40
fr.no.
Fig. 2. Elution profiles of the pure Klebsiella enzymes after analytical chromatofocusing on columns of PBE 94. Samples of known activities and protein contents were taken from the peak fractions, shown in Fig. la-c for the enzymes of K. pneumoniae Kl SC 10436, K. aerogenes Kl 1082 E and K. oxytoca 1082 E, respectively. The K. oxytoca 20 (d) enzyme sample was taken from the classically purified main fraction. The starting buffers, pH 7.4 for a-c and pH 9.4 for d, respectively, and Polybuffer 74, pH 7.4 for a-c and Polybuffer 94, pH 9.4 for d, respectively, were as described in the Experimental. The amount of each enzyme is given in Results and Discussion. Elution conditions: pH (-) and activity (- - -) monitoring as described in the Experimental; protein (, .) is shown only for K. oxytoca 20.
dation rather than binding had occurred; however, a further p-lactamase fraction was eluted with borate buffer. Because of the low affinity of the enzyme towards the hydrophobic Type B column, an hydrophilic Type L phenylboronic acid-agarose affinity material or a simple classical prepurification method could be used. These purified enzymes were analyzed by chromatofocusing in comparison with the original crude extracts. We wanted to learn whether this technique is suitable not only for the determination of the pI values of the constitutive cephalosporinases of different Klebsiella strains, but also whether it can be used as an alternative for preparation purposes by exploiting the differences in pI of various accompanying proteins. The results of analytical chromatofocusing experiments with the highly purified enzymes are shown in Fig. 2. The peak fraction 21 of K. pneumoniae Kl SC 10436 had pI = 6.4 (Fig. 2a). Of the 15.5 units applied to the column, 13.4 units, X6%, were eluted. About one third of the activity was found in fractions 16-20, which may reflect partial separation of minor satellite enzymes of higher pI values, known to exist in KZebsiella fi-lactamase preparations. However, AIEF data are not available for comparison. K. aerogenes Kl 1082 E cephalosporinase was essentially eluted in two fractions (Fig. 2b).
S. GAL et al.
222 OD,
(al
PH
u*ir.-’ 7.5.
5-
2.5.
ODzm
PH
0.5
0.25
10
20
30
40
fr.m.
10
20
30
fr.
no.
Fig. 3. Elution profiles of the crude Klebsida enzymes after analytical chromatofocusing on columns of PBE 94. Samples of known activities and protein contents were applied to the columns as given in Results and Discussion. Starting buffers and PolybuKers as described in the Experimental and in Fig. 2. Elution conditions: pH (), protein (. . .) and activity (- - -) monitoring as described in the Experimental. Enzymes as in Fig. 1.
The enzyme showed pl = 6.505 in fraction 18 and p1 = 6.495 in fraction 19. Of the 8.6 units applied, 8.5 units, 99%, were eluted. AIEF gave identical pI values21. The pI value of the K. oxytoca 1082 E enzyme in fraction 18 was 6.42 (pl = 6.4-6.6 by AIEFz5). Elution resulted in the recovery of 6.7 units, 93%, of the total of 7.2 units applied to the column (Fig. 2~). The analysis of the K. oxytoca 20 enzyme is shown in Fig. 2d. The peak fractions 15-19 represented 0.7 units, 90%, of the 0.77 units applied to the column. Fraction 17 had pl = 7.62 (PI = 7.67.9 by AIEFzs). Analytical chromatofocusing of the crude enzymes is shown in Fig. 3. Of 18.2 units of crude K. pneumoniue Kl SC 10436 applied, 17.8 units, 98%, were recovered (Fig. 3a). Peak fraction 16 had pZ = 6.42. The symmetrical profile may suggest the presence of minor satellite enzyme species at both sides of the main activity fraction. For crude K. aerogenes Kl 1082 E enzyme, 24.3 units were applied of which 21.0 units were eluted (Fig. 3b), 86%. The pI value of fraction 17 was 6.5, identical to that obtained with AIEFzl. Fraction 15 of the K. oxytoca 1082 E enzyme had pI = 6.4. Of the 73.8 units applied, 67.2 units, 91%, were eluted (Fig. 3~). Of the 1.68 units of K. oxytoca 20 enzyme applied to the column, 1.53 units, 90%, were recovered (Fig. 3d). Fractions 16 and 17 had pZ = 7.6 and 7.56, respectively. Fig. 4 illustrates the results of SDS-PAGE of the crude enzymes and of those prefractionated by phenylboronic acid-agarose affinity column in the cases of K. pneumoniae Kl SC 10436 K. aerogenes Kl 1082 E and K. oxytoca 1082 E and by the
CHROMATOFOCUSING
OF /I-LACTAMASES.
VIII.
223
Fig. 4. SDS-PAGE of crude and purified Klebsiella enzymes. K. pneumoniae Kl SC 10436 (a), K. aerogenes Kl 1082 E (b) and K. oxytoca 1082 E (c) were purified by phenylboronic acid-agarose and K. oxytoca 20 (d) by a classical molecular sieving method as described in the Experimental. Portions of 5 ~1 of the crude and 50 ~1 of the purified enzymes from the peak chromatographic fractions were applied to each track and the gels were subjected to 90 V, 40 mA, for 3.5 h. Protein was detected with Coomassie Brilliant Blue R250. The anode was at the bottom of the gels. The samples were run simultaneously in the same apparatus.
classical method in the case of K. oxytoca 20. Except for K. oxytoca 20 cephalosporinase, the main enzyme fractions were essentially 100% pure. The basic goal, excellent concentration of the enzymes and elimination of the accompanying proteins, was thus achieved. Although the p1data from AIEF and from chromatofocusing are not identical, pl values obtained by our method are ranging at the lower levels of the AIEF pl intervals published for K. oxytoca 1082 E2 5 and K. oxytoca 202 5, respectively, the pl values for the K. aerogenes Kl 1082 E enzyme obtained by the two methods were the same. REFERENCES T. Sawai, M. Kanno and K. Tsukamoto, J. Bacterial., 152 (1982) 567. R. Labia, A. Morand, M. Guionie, M. Heitz and J. S. Pitton, Pathol. Biol., 34 (1986) 611. J. S. Pitton, M. Heitz and R. Labia, Current Chemotherapy, Am. Sot. Microbial., Washington, DC, 1978, pp. 482484. R. Labia, A. Morand and A. Kazmierczak, J. Antimicrob. Chemother., 14 Suppl. B (1984) 45. C. A. Hart and A. Percival, J. Antimicrob. Chemother., 9 (1982) 275. L. R. Then, M. P. Glauser, P. Angher and M. Arishawa, Zbl. Bakt. Hyg., I. Abt. Orig. A, 254 (1983) 469. M. H. Richmond, J. Antimicrob. Chemother.. 6 (1980) 445. A. M. Philippon, G. C. Paul and P. A. Nevot, Proc. 1 Ith International Congress of Chemotherapy and 19th Conference on Antimicrobial Agents and Chemotherapy, Am. Sot. Microbial., Washington, DC, 1980, pp. 327-329.
224
S. GAL et al.
9 R. Labia, C. Fabre, J. M. Masson, M. Barthelemy, M. Heitz and J. S. Pitton, J. Antimicrob. Chemo(her., 5 (1979) 375. 10 J. S. Pitton, M. Heitz and R. Labia. in W. Siegenthaler and R. Liithy (Editors), Current Chemolher., Proc. 10th ICC, Zurich, 1977, Am. Sot. Microbial., Washington DC, 1978, pp. 482484. 11 T. Sawai, S. Yamagishi and S. Mitsuhashi, J. Bacterial., 115 (1973) 1045. 12 P. M. Shap and W. Stille, J. Antimicrob. Chemother., 11 (1983) 597. 13 M. E. Nugent and R. W. Hedges, Mol. Gen. Genet., 175 (1979) 239. 14 Y. Arakawa, M. Ohta, N. Kido, Y. Fujii, T. Komatsu and N. Kato, FEBS Left., 207 (1986) 69. 15 E. Cornel, A. Philippon, G. Paul and M. Guenounou, Ann. Microbial. (Inst. Pasteur), 133 B (1982) 365. 16 W. Cullmann, W. Opferkuch, M. Steiglitz and W. Dick, Chemother., 30 (1984) 175. 17 A. Philippon and G. Paul, Mkdecine et Maladies Znfectieuses, 11 (1986) 644. 18 R. W. Hedges and A. E. Jacob, Mol. Gen. Genet., 132 (1974) 31. 19 J. S. Pitton, Ergeb. Physiol., Biol. Chem. Exp. Pharmacol., 65 (1972) 15-93. 20 B. L. Toth-Martinez, S- Gal, F. J. Hemadi, L. Kiss and P. Nat&i, J. Chromatogr., 287 (1984) 413. 21 S. J. Cartwright and S. G. Waley, Biochem. J., 221 (1984) 505. 22 C. H. O’Callaghan, A. Morris, S. Kirby and A. H. Shingler, Anfimicrob. Agents Chemother., 1 (1972) 283. 23 0. H. Lowry, N. J. Rosenbrough, A. L. Farr and R. J. Randall, J. Biol. Chem., 193 (1951) 265. 24 K. Weber and M. Osborn, Antimicrob. Agents Chemother., 5 (1969) 25. 25 L. R. Then, personal communication.
403 (1987) 217-224 B.V., Amsterdam -
Printed
in The Netherlands
19 586
USE OF CHROMATOFOCUSING VIII*. ANALYTICAL PHALOSPORINASES
OF b-LACTAMASES
CHROMATOFOCUSING OF CHROMOSOMAL FROM FOUR KLEBSIELLA STRAINS
SUSANNE GAL, ANDREA TAR, NEZ* and FERENC J. HERNADI Chemotherapy Section, H-4012 (Hungary)
FOR SEPARATION
Department
MAGDOLNA of Pharmacology,
FROMMER-FILEP, University
Medical
BELA
CE-
L. TOTH-MARTI-
School of Dehrecen,
Debrecen
and LkZLt)
KISS
Department (Received
of Biochemistry, March
28th,
L. Kossuth
University,
Debrecen
H-4010
(Hungary)
1987)
SUMMARY
Although still there are Klebsiella strains which do not harbour plasmids and produce constitutive chromosomal p-lactamases, recently clinical isolates were found in ever increasing numbers carrying mainly TEM-, CARB- and OXA type R-factors. We selected four chromosomal cephalosporinase producing Klebsiella strains to study the p1 values of the enzymes and their simultaneous separability from accompanying proteins by chromatofocusing techniques. We compared pl values of the pure and the crude preparations: K. pneumoniae Kl SC 10436: pl,,,, = 6.4, 6.42; K. aerogenes Kl 1082 E: ~1~,,,, = 6.5, pz_& = 6.5; K. oxytoca 1082 p&rude = = 6.42, PIcrude = 6.4; K. oxytoca 20: pl,,,, = 7.62, pIcrude = 7.6. Excellent E: P& agreement of the p1 values among each other, but occasional differences with those obtained by analytical isoelectrofocusing are attributed to methodological diversities and to the presence of satellite enzymes, known to exist in Klebsiella.
INTRODUCTION
Klebsiella is one of the most commonly distributed bacterium species all over the world. Many Klebsieila strains do not harbour plasmids and produce constitutive chromosomal /%lactamases. Most of these /?-lactamases preferentially hydrolyze cephalosporins, including many of those that are resistant to plasmid-coded enzymes. Sawai et al.’ distinguished two major groups of Klebsiella enzymes of broad substrate specificity: (1) chromosomally coded cephalosporinases that hydrolyze benzylpenil For Part VII, see A. Tar, S. Gil, B. L. Toth-Martinez, 368 (1986) 427.
F. J. Hernhdi
0021-9673/87/$03.50
B.V.
0
1987 Elsevier
Science Publishers
and L. Kiss, J. Chromatogr.,
218
S. GAL et al.
cillin, ampicillin and carbenicillin as well, e.g., K. oxytoca FOU 1, FOU 2, FOU 3, FOU 13, pZ = 5.52; R 700, pZ = 5.72; R 30, pZ = 5.82 or 6.03,4; FOU 11, pZ = 6.02; FOU 12, R 210, D 488, pZ = 6.32; 3859, pZ = 6.65; 22264, pZ between 7.9 and 8.26; K. aerogenes K 1, pZ = 6.657; K. pneumoniae 1103, pZ = 7.12~s; (2) chromosomally determined typical cephalosporinases that have little or no activity against the above penicillins and have pZ values above 7.06,9+12. Generally speaking none of these enzymes is identical with the plasmid-mediated fi-lactamases, with one exception’j that is indistinguishable by kinetic analysis and isoelectric focusing from the plasmid SHV-1 p-lactamase (TEM-1 Type 2 or PIT 2, pZ = 7.7; sulfhydryl variable, Temoniera is the name of a patient and PIT is from Professor J. S. Pitton). This is the variant of SHV-1 incorporated into a non-transmissible replicon, probably the bacterial chromosome. Nugent and Hedges13 proposed that the SHV-1 B-lactamase gene evolved as a chromosomal gene in Klebsiella spp. and was later incorporated into a plasmid. Recently Arakawa et al. I4 found that the chromosomally encoded fi-lactamase gene of K. pneumoniae LEN-l (03:Kl) and the TEM_enzyme of Tn3 (transposon) show 67% homology of the deduced amino acid sequence, suggesting that the latter originates from K. pneumoniae. Recently, clinical K. pneumoniae and mainly K. oxytoca isolates have been found to show different modality of resistance related to their broad-spectrum /$ lactamases. These enzymes are produced in high yield by these Klebsiella strains together with more or less chromosomal enzymes 2,5,6. No doubt, these strains harbour different plasmids as well, solely or in combination as reported in some cases: K. oxytoca 22812, 22820, 23659 harbour TEM-1, pZ = 5.4 with high chromosomal enzyme of pZ7.9-8.26; K. pneumoniae S6 and S7 have SHV-1 and SHV-1 + TEM-1 /?-lactamases, respectively15, 22505 has enzymes of pZ 6.5, 7.2, 7.6 and 8.1 and TEM-1 6. HMS-1 of pZ = 5.0 was also found in Klebsieh strains16. As far as this third group is concerned, the following multiple extrachromosomal #?-lactamases have been reported in various Klebsielia strains’ 7: TEM-1, TEM-2, pI = 5.6; CARB-1, pZ 2 5.3; CARB-2, pZ = 5.7; CARB-3, pZ = 5.75; CARB-4, pZ = 4.3; OXA-1, pZ = 7.4; OXA-2, pZ of 7.4 alid 7.7; OXA-3, pZ = 7.0; OXA-4, pZ between 7.45 and 7.5; OXA-5, pZ = 7.6; OXA-6, pZ = 7.8; OXA-7, pZ = 7.65. Another characteristic of these enzymes is an extended spectrum to new plactams of the third generation and azthreonam by mutationle. This was encountered in K. orizae: adaptation of the plasmid SHV-1 resulted in the form&on of SHV-2, pZ = 7.7 (unchanged). Plasmid P45319 carries the transposon coding for SHV-113; the mutant plasmid, however, is called pBP6017. The isoelectric points of enzymes from four Klebsiella strains, producing constitutive chromosomal fi-lactamases in high yield, were examined in our analytical chromatofocusing system and compared with those obtained by analytical isoelectrofocusing (AIEF). The enzymes were also characterized by their separability from accompanying proteins. EXPERIMENTAL
Bacterial strains
Constitutively producing strains of K. oxytoca 1082 E and 20 were kindly provided by Dr. L. R. Then, Department of Pharmaceutical Research, F. Hoff-
CHROMATOFOCUSING
OF b-LACTAMASES.
VIII.
219
mann-La Roche, Basle, Switzerland, K. oxyloca 20 by Dr. C. E. Cooper, Chemotherapeutic Research Centre, Brockham Park, Betchworth, U.K., K. aerogenes Kl 1082 E by Dr. I. N. Simpson, Glaxo Group Research, Greenford, U.K. and by Dr. S. G. Waley, Sir William Dunn School of Pathology, University of Oxford, Oxford, U.K. K. pneumoniae Kl SC 10436 by Dr. K. Bush, The Squibb Institute for Medical Research, Princeton, NJ, U.S.A. Details of the culturing procedure have been published elsewhereZo. PuriJication of B-lactamases Except for K. oxytoca 20, the p-lactamases were purified by affinity chromatography on phenylboronic acid-agarose (hydrophobic spacer arm, Type B) according to Cartwright and Waley 21. The K. oxytoca 20 fi-lactamase showed an unusual elution profile (Fig. Id) on the affinity column: a major fraction was eluted first with 20 mM triethanolamine hydrochloride-0.5 A4 sodium chloride, pH 7.0 washing buffer and another fraction was obtained with 0.5 M borate-0.5 M sodium chloride, pH 7.0 elution buffer. This was an indication that the Type B affinity material was not suitable for quantitative purification of this enzyme. A similar retarded elution was encountered by Cartwright and Waley2’ in the case of K. aerogenes Kl 1082 E on an hydrophylic Type L boronic acid column. We used a classical prepurification method for the K. oxytoca 20 enzyme: the crude enzyme was applied to Sephadex G-75 (20 cm x 3.5 cm) and eluted by 0.05 M phosphate, pH 7.0; after pooling, the active fractions were chromatographed on QAE-Sephadex A-50 (20 cm x 3.5 cm) using 0.025 M Tris-HCl buffer, pH 7.5. A portion of this purified enzyme was used for chromatofocusing as described below. Crude enzyme fractions were made by freezing-thawing and by sonication of resuspended washed cells in 0.05 M phosphate buffer, pH 7.0. Following treatments with nuclease, supernatants were prepared by ultracentrifugation at 105 000 g for 1 h and extensively dialyzed against the above buffer. Chromatofocusing, enzyme assay, protein determination and sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) A Pharmacia C lo/20 column (20 cm x 1 cm) packed with PBE 94 chromatofocusing material was equilibrated with 0.025 A4 imidazole hydrochloride buffer, pH 7.4 for K. pneumoniae Kl SC 10436, K. aerogenes Kl 1082 E and K. oxytoca 1082 E strains or with 0.025 M ethanolamine-acetic acid buffer, pH 9.4 for K. oxytoca 20. For the first three cephalosporinases, the eluent was Polybuffer 74 diluted eight-fold in distilled water and adjusted to pH 4.0 with hydrochloric acid, in the case of K. oxytoca 20, the eluent was Polybuffer 96 diluted ten-fold in distilled water and adjusted to pH 6.0 with acetic acid. fi-Lactamase activity was assayed by measuring the absorbance of Nitrocefin at 486 nm in a cell of 1 cm pathlength as described by O’Callaghan et aZ.22.One unit is that amount of enzyme which is able to hydrolyze 1 ymol of Nitrocefin in 1 min at 37°C. The protein content of the fractions was estimated by measuring their absorbance at 280 nm or as described by Lowry et a1.23. SDS-PAGE was performed according to Weber and 0sborn24.
S. GAL (a)
U.fr.-’
1
1
U.fr._’
CC)
OD2ao
er al.
0.30
(b)
ODzso
Cd)
\; / 0.10 -
M.
J, I c
/
- 0.25
/I I
5
i0
15
fr. no.
Fig. 1. Elution profiles of the crude extracts of /?-lactamases of four Klebsiella strains on Type B phenylboronic acid-agarose affinity columns. Crude extract samples, dialyzed against 20 mM triethanolamine hydrochloride-O.5 M sodium chloride, pH 7.0, were applied; the columns were washed with the same buffer until protein-free fractions appeared. The enzymes were eluted with 0.5 M berate-0.5 M sodium chloride, pH 7.0, elution buffer. Protein (. . .) and activity (- - -) monitoring were as described in the Experimental. The amount of each enzyme and the volumes of the loading buffer are given in Results and Discussion. Fractions eluted were 1.5 ml. (a) K. pneumoniue K1 SC 10436; (b) K. aerogenes Kl 1082 E; (c) K. oxytoca 1082 E; (d) K. oxytoca 20. fr.no. = Fraction number; U.fr.-’ = units per fraction.
RESULTS
AND
DISCUSSION
Fig. 1 shows the results of the purification of Klehsiella p-lactamases on Type B phenylboronic acid-agarose affinity material. K. pneumoniae Kl SC 10436 (a), K. aerogenes Kl 1082 E (b) and K. oxytoca 1082 E (c) had low protein contents and high activities, while K. oxytoca 20 had about the same protein content but low activity (d). When 128 units of K. pneumoniae Kl SC 10436 crude enzyme were loaded onto a column (3 cm x 0.9 cm) in 1.5 ml loading buffer, no detectable amount of activity passed through the column either on loading or upon washing (Fig. la). The enzyme was eluted in two fractions of 3 ml, yielding 121 units or 95% recovery. Under similar conditions, 165 units of crude K. aerogenes Kl 1082 E enzyme were loaded and 161 units, i.e., 98% recovery, were eluted in two fractions of 3 ml (Fig. 1.b). Of 115.5 units of crude K. oxytoca 1082 E enzyme loaded 112.5 units, 96% recovery, were eluted in 3 ml (Fig. lc). The high purification capacity of the Type B affinity material with the above enzymes is noteworthy. When 5.4 units of crude K. oxytoca 20 (Fig. Id) enzyme in 6 ml were loaded on the affinity column, only 4.3 units, 80%, were eluted. The washing buffer eluted about three quarters of the enzyme in the first ten eluate fractions, i.e., clearly retar-
CHROMATOFOCUSING
10
20
OF p-LACTAMASES.
30 (b)
lb
40 U.fr._’
A II 1I I I
5.
4 0480
i0
I
i0 Cd)
pH
4.0 U*fr.-’
- 0.3
I
I
I I
I I
I:\
VIII.
I
’
;
I
I
’
- 0.2
. 0.1 I I
IO
20
30
40
fr.no.
.^
1”
I 1
zu
. .. .. . . . . . .. .
30
40
fr.no.
Fig. 2. Elution profiles of the pure Klebsiella enzymes after analytical chromatofocusing on columns of PBE 94. Samples of known activities and protein contents were taken from the peak fractions, shown in Fig. la-c for the enzymes of K. pneumoniae Kl SC 10436, K. aerogenes Kl 1082 E and K. oxytoca 1082 E, respectively. The K. oxytoca 20 (d) enzyme sample was taken from the classically purified main fraction. The starting buffers, pH 7.4 for a-c and pH 9.4 for d, respectively, and Polybuffer 74, pH 7.4 for a-c and Polybuffer 94, pH 9.4 for d, respectively, were as described in the Experimental. The amount of each enzyme is given in Results and Discussion. Elution conditions: pH (-) and activity (- - -) monitoring as described in the Experimental; protein (, .) is shown only for K. oxytoca 20.
dation rather than binding had occurred; however, a further p-lactamase fraction was eluted with borate buffer. Because of the low affinity of the enzyme towards the hydrophobic Type B column, an hydrophilic Type L phenylboronic acid-agarose affinity material or a simple classical prepurification method could be used. These purified enzymes were analyzed by chromatofocusing in comparison with the original crude extracts. We wanted to learn whether this technique is suitable not only for the determination of the pI values of the constitutive cephalosporinases of different Klebsiella strains, but also whether it can be used as an alternative for preparation purposes by exploiting the differences in pI of various accompanying proteins. The results of analytical chromatofocusing experiments with the highly purified enzymes are shown in Fig. 2. The peak fraction 21 of K. pneumoniae Kl SC 10436 had pI = 6.4 (Fig. 2a). Of the 15.5 units applied to the column, 13.4 units, X6%, were eluted. About one third of the activity was found in fractions 16-20, which may reflect partial separation of minor satellite enzymes of higher pI values, known to exist in KZebsiella fi-lactamase preparations. However, AIEF data are not available for comparison. K. aerogenes Kl 1082 E cephalosporinase was essentially eluted in two fractions (Fig. 2b).
S. GAL et al.
222 OD,
(al
PH
u*ir.-’ 7.5.
5-
2.5.
ODzm
PH
0.5
0.25
10
20
30
40
fr.m.
10
20
30
fr.
no.
Fig. 3. Elution profiles of the crude Klebsida enzymes after analytical chromatofocusing on columns of PBE 94. Samples of known activities and protein contents were applied to the columns as given in Results and Discussion. Starting buffers and PolybuKers as described in the Experimental and in Fig. 2. Elution conditions: pH (), protein (. . .) and activity (- - -) monitoring as described in the Experimental. Enzymes as in Fig. 1.
The enzyme showed pl = 6.505 in fraction 18 and p1 = 6.495 in fraction 19. Of the 8.6 units applied, 8.5 units, 99%, were eluted. AIEF gave identical pI values21. The pI value of the K. oxytoca 1082 E enzyme in fraction 18 was 6.42 (pl = 6.4-6.6 by AIEFz5). Elution resulted in the recovery of 6.7 units, 93%, of the total of 7.2 units applied to the column (Fig. 2~). The analysis of the K. oxytoca 20 enzyme is shown in Fig. 2d. The peak fractions 15-19 represented 0.7 units, 90%, of the 0.77 units applied to the column. Fraction 17 had pl = 7.62 (PI = 7.67.9 by AIEFzs). Analytical chromatofocusing of the crude enzymes is shown in Fig. 3. Of 18.2 units of crude K. pneumoniue Kl SC 10436 applied, 17.8 units, 98%, were recovered (Fig. 3a). Peak fraction 16 had pZ = 6.42. The symmetrical profile may suggest the presence of minor satellite enzyme species at both sides of the main activity fraction. For crude K. aerogenes Kl 1082 E enzyme, 24.3 units were applied of which 21.0 units were eluted (Fig. 3b), 86%. The pI value of fraction 17 was 6.5, identical to that obtained with AIEFzl. Fraction 15 of the K. oxytoca 1082 E enzyme had pI = 6.4. Of the 73.8 units applied, 67.2 units, 91%, were eluted (Fig. 3~). Of the 1.68 units of K. oxytoca 20 enzyme applied to the column, 1.53 units, 90%, were recovered (Fig. 3d). Fractions 16 and 17 had pZ = 7.6 and 7.56, respectively. Fig. 4 illustrates the results of SDS-PAGE of the crude enzymes and of those prefractionated by phenylboronic acid-agarose affinity column in the cases of K. pneumoniae Kl SC 10436 K. aerogenes Kl 1082 E and K. oxytoca 1082 E and by the
CHROMATOFOCUSING
OF /I-LACTAMASES.
VIII.
223
Fig. 4. SDS-PAGE of crude and purified Klebsiella enzymes. K. pneumoniae Kl SC 10436 (a), K. aerogenes Kl 1082 E (b) and K. oxytoca 1082 E (c) were purified by phenylboronic acid-agarose and K. oxytoca 20 (d) by a classical molecular sieving method as described in the Experimental. Portions of 5 ~1 of the crude and 50 ~1 of the purified enzymes from the peak chromatographic fractions were applied to each track and the gels were subjected to 90 V, 40 mA, for 3.5 h. Protein was detected with Coomassie Brilliant Blue R250. The anode was at the bottom of the gels. The samples were run simultaneously in the same apparatus.
classical method in the case of K. oxytoca 20. Except for K. oxytoca 20 cephalosporinase, the main enzyme fractions were essentially 100% pure. The basic goal, excellent concentration of the enzymes and elimination of the accompanying proteins, was thus achieved. Although the p1data from AIEF and from chromatofocusing are not identical, pl values obtained by our method are ranging at the lower levels of the AIEF pl intervals published for K. oxytoca 1082 E2 5 and K. oxytoca 202 5, respectively, the pl values for the K. aerogenes Kl 1082 E enzyme obtained by the two methods were the same. REFERENCES T. Sawai, M. Kanno and K. Tsukamoto, J. Bacterial., 152 (1982) 567. R. Labia, A. Morand, M. Guionie, M. Heitz and J. S. Pitton, Pathol. Biol., 34 (1986) 611. J. S. Pitton, M. Heitz and R. Labia, Current Chemotherapy, Am. Sot. Microbial., Washington, DC, 1978, pp. 482484. R. Labia, A. Morand and A. Kazmierczak, J. Antimicrob. Chemother., 14 Suppl. B (1984) 45. C. A. Hart and A. Percival, J. Antimicrob. Chemother., 9 (1982) 275. L. R. Then, M. P. Glauser, P. Angher and M. Arishawa, Zbl. Bakt. Hyg., I. Abt. Orig. A, 254 (1983) 469. M. H. Richmond, J. Antimicrob. Chemother.. 6 (1980) 445. A. M. Philippon, G. C. Paul and P. A. Nevot, Proc. 1 Ith International Congress of Chemotherapy and 19th Conference on Antimicrobial Agents and Chemotherapy, Am. Sot. Microbial., Washington, DC, 1980, pp. 327-329.
224
S. GAL et al.
9 R. Labia, C. Fabre, J. M. Masson, M. Barthelemy, M. Heitz and J. S. Pitton, J. Antimicrob. Chemo(her., 5 (1979) 375. 10 J. S. Pitton, M. Heitz and R. Labia. in W. Siegenthaler and R. Liithy (Editors), Current Chemolher., Proc. 10th ICC, Zurich, 1977, Am. Sot. Microbial., Washington DC, 1978, pp. 482484. 11 T. Sawai, S. Yamagishi and S. Mitsuhashi, J. Bacterial., 115 (1973) 1045. 12 P. M. Shap and W. Stille, J. Antimicrob. Chemother., 11 (1983) 597. 13 M. E. Nugent and R. W. Hedges, Mol. Gen. Genet., 175 (1979) 239. 14 Y. Arakawa, M. Ohta, N. Kido, Y. Fujii, T. Komatsu and N. Kato, FEBS Left., 207 (1986) 69. 15 E. Cornel, A. Philippon, G. Paul and M. Guenounou, Ann. Microbial. (Inst. Pasteur), 133 B (1982) 365. 16 W. Cullmann, W. Opferkuch, M. Steiglitz and W. Dick, Chemother., 30 (1984) 175. 17 A. Philippon and G. Paul, Mkdecine et Maladies Znfectieuses, 11 (1986) 644. 18 R. W. Hedges and A. E. Jacob, Mol. Gen. Genet., 132 (1974) 31. 19 J. S. Pitton, Ergeb. Physiol., Biol. Chem. Exp. Pharmacol., 65 (1972) 15-93. 20 B. L. Toth-Martinez, S- Gal, F. J. Hemadi, L. Kiss and P. Nat&i, J. Chromatogr., 287 (1984) 413. 21 S. J. Cartwright and S. G. Waley, Biochem. J., 221 (1984) 505. 22 C. H. O’Callaghan, A. Morris, S. Kirby and A. H. Shingler, Anfimicrob. Agents Chemother., 1 (1972) 283. 23 0. H. Lowry, N. J. Rosenbrough, A. L. Farr and R. J. Randall, J. Biol. Chem., 193 (1951) 265. 24 K. Weber and M. Osborn, Antimicrob. Agents Chemother., 5 (1969) 25. 25 L. R. Then, personal communication.